Everyone makes mistakes in thinking. Some are random: a moment of distraction, a misread number, a slip of the tongue. A cognitive bias is different. It is a systematic, predictable pattern of error in judgment, a tendency that pushes many people's thinking in the same direction in the same kinds of situations. Because biases are predictable, they can be studied, and we can learn to design around them.
Heuristics: mental shortcuts
Most biases grow out of heuristics, the mental shortcuts or rules of thumb the mind uses to make quick judgments with limited information. “If a restaurant is full, the food is probably good.” “If I can easily think of examples, it must be common.” Shortcuts like these usually work well enough and save enormous effort. A bias appears when a shortcut is used in a situation it does not fit.
Why the mind uses shortcuts
In the 1950s the economist and psychologist Herbert Simon argued that real people have bounded rationality: they decide with limited time, information and mental capacity, so they cannot weigh every option perfectly. Instead they often “satisfice,” his word for choosing the first option that is good enough rather than searching endlessly for the best.
Many researchers add an evolutionary explanation. Our ancestors had to make fast decisions with high stakes. Treating a rustle in the grass as a predator when it was only the wind cost little; making the opposite mistake could cost a life. A mind tuned for speed and survival was not built to read statistical tables. Such evolutionary stories are plausible but hard to test, so treat them as likely explanations rather than settled history.
Two modes of thinking
The psychologist Daniel Kahneman popularized a helpful picture in his 2011 book Thinking, Fast and Slow, using labels introduced earlier by the researchers Keith Stanovich and Richard West.
- System 1 is fast, automatic and intuitive. It recognizes a friend's face, hears anger in a voice, and completes “bread and ...” without effort.
- System 2 is slow, deliberate and effortful. It multiplies 17 by 24, fills in a tax form, or checks whether an argument really holds together.
Kahneman stressed that the two systems are a useful metaphor, not two separate places in the brain. Usually System 1 runs the show and System 2 simply accepts its suggestions. Many biases occur when System 1 offers a quick, plausible answer and System 2 never checks it.
A famous puzzle shows the pattern. A bat and a ball cost $1.10 together. The bat costs $1.00 more than the ball. How much does the ball cost? Most people's first thought is 10 cents. But then the bat would cost $1.10 and the pair $1.20. The correct answer is 5 cents: a ball at $0.05 plus a bat at $1.05 makes $1.10.
The 1974 paper
In 1974 Amos Tversky and Daniel Kahneman published “Judgment under Uncertainty: Heuristics and Biases” in the journal Science. It described three heuristics: representativeness, availability, and anchoring and adjustment (Level 4 explains each). It showed that these shortcuts produce predictable errors, even among people trained in statistics. In 2002 Kahneman received the Nobel Memorial Prize in Economic Sciences for this work; Tversky had died in 1996.
Biases are not stupidity
Intelligent, educated people show biases too. Worse, knowing about a bias does not automatically protect you. People readily spot biases in others while seeing themselves as objective, a pattern researchers call the bias blind spot. That is why the final level of this course focuses on habits and procedures, not just awareness.
Worked example: two more quick puzzles
The bat-and-ball puzzle comes from a short test devised by the psychologist Shane Frederick in 2005. Try two more from the same family before reading on.
- Machines. If 5 machines take 5 minutes to make 5 widgets, how long would 100 machines take to make 100 widgets? The answer that leaps to mind is 100 minutes. Now check it: each machine takes 5 minutes to make one widget. With 100 machines working at once, 100 widgets take 5 minutes.
- Lily pads. A patch of lily pads doubles in size every day. It takes 48 days to cover the whole lake. How long does it take to cover half the lake? The tempting answer is 24 days. But if the patch doubles daily, it must have been half the lake the day before it was full: 47 days.
Notice what went wrong and what fixed it. Almost everyone who gets these wrong gives the same wrong answer, the signature of a bias rather than a slip. The arithmetic is easy. What made the difference was a habit: pausing to ask “Does my answer actually satisfy the conditions?” That is System 2 doing its checking job.
One caution: the lesson is not that intuition is useless. In familiar situations with quick feedback, such as an experienced nurse noticing that a patient looks wrong, intuition can be remarkably accurate. The skill is knowing when a quick answer deserves a second look.